1) Molecular mechanisms of treatment response in rectal cancer The aim of this study is to uncover novel mechanisms of radio-resistance in colorectal cancer cells. To this end, we analyzed two cancer cell lines, SW480 (Colon) and SW1463 (Rectum), both mimicking the typical chromosomal aberration- and gene expression patterns characteristic of colorectal carcinomas. To increase the radiation resistance of these cells, either by selecting a priori resistant clones or induction of de novo mutations, we repetitively irradiated single cell derived populations of the two cell lines with 40 doses of 2 GY over a period of 40 passages. This led to the establishment of 9 radiation resistant, single-cell derived, populations per cell line. Furthermore we established 3 independent, age matched, mock treated cell populations per cell line to serve as controls. Using this approach we could establish 9 single cell derived radiation resistant clones as well as 3 age-matched controls each for two well established and characterized cell lines. We first analyzed copy number variations using arrayCGH. We found several aberrations unique to the resistant clones. Using RNAseq we are aiming to identify the underlying mechanisms of radiation resistance, including changes in the transcriptome, gene expression as well as alternative splicing. Furthermore, we are expecting several de novo mutations possibly associated with radiation resistance. Out of the 9 radiation resistant clones of SW1463, we picked 4 showing the largest increase in radiation resistance (based on colony-formation assays) for RNAseq. To understand the difference in the response of the resistant clones compared to the controls to irradiation treatment we isolated two sets of RNA from cells (3 controls and 4 resistant clones). The first and the second set of RNA are derived from the same flask to keep variance in the genetic background as low as possible. The first set was isolated 6 h after irradiation with a total dose of 4 Gy, the second was isolated at the same time point but without irradiation. After RNAseq we first want to validate the new targets using qRT-PCR in SW1463 and SW480. Targets being found in both cell lines to be upregulated in resistant clones will be functionally validated by RNAi (siRNA) based loss-of-function experiments. Genes downregulated in resistant clones will be re-expressed. Genes/SNPs/deregulated pathways shown to modulate radiation response will be further analyzed, we are planning to validate changes in protein expression by Western blotting and analysis of underlying changes in DNA damage repair capacity using FACS based H2AX-staining and reporter assays measuring the NHEJ and MMEJ. Due to a very close cooperation with the Department of Surgery, Goettingen, we have access to gene expression profiles of 260 pretreatment primary rectal cancers and SNP-profiles of 80 rectal cancer patients as well as their clinical data, enabling us to validate our findings in vivo. After in silico validation we will analyze a subset of our targets using qRT-PCR and IHC in primary tumors. 2) The impact of genomic heterogeneity in colorectal carcinoma on therapy response - a study of monoclonal primary colorectal cell lines Individual response to chemotherapy in colorectal cancer patients is highly variable and the underlying mechanisms of treatment resistance of cancer cells are poorly understood. Recent studies revealed a considerable degree of genomic tumor heterogeneity. We hypothesize that this heterogeneity has a direct impact on treatment response as subpopulations of cancer cells are resistant to currently used chemotherapeutics and facilitate tumor growth under treatment. To address this highly relevant clinical issue, patient-derived primary colorectal cell lines were established. The tissue was derived from a 59 year old male presenting with a low grade adenocarcinoma of the rectosigmoid colon (ypT3, pN0) which was implanted heterotopically into animmunodeficient nude mouse. After in vivo growth, the tumor was dissociated and introduced to an in vitro culture where invaded murine stromal cells were depleted using antibody-based columns. The cell line proved to be of human origin by immunofluorescence for human Keratin 20 and EpCAM as well as genotyping by short tandem repeat profiling. The characterization of the genome by array CGH and spectral karyotyping (SKY) revealed a highly complex hyperdiploid karyotype with numerous chromosomal imbalances, which are typically found in colorectal cancers. In addition to the typical pattern of gains found in chromosomes 7, 13 and 20, other structural abnormalities were discovered including an isochromosome 1q. Multiplex-FISH by consecutive hybridization of probes for 15 gene loci, which are known to be relevant in colorectal cancer genesis, revealed a significant degree of clonal heterogeneity of the cell line. Single cell sorting was then performed employing limited dilution to establish single cell derived cell lines from the genomically well characterized parental cell line. Analyses of the genome and transcriptome of single cell derived cell lines and exposure of the cells to clinically relevant doses of 5-FU and oxaliplatin will ensue. This will allow the comparison of the respective chemotherapeutic sensitivities to the identified aberration profiles. These data will facilitate the understanding of therapy resistance and allow a reliable prediction of the patient's response to the employed chemotherapeutic drug. A tailored chemotherapy is an important step towards individualized treatment in colorectal cancer patients to avoid therapy resistance. 3) Comprehensive analysis of the role of WNT-signaling in CRC tumorigenesis and treatment response The organization of the interphase nucleus reflects a dynamical interaction between 3D genome structure and function and its relationship to phenotype, a concept known as the 4D Nucleome (4DN). 4DN research requires a comprehensive view of architecture, gene expression, the proteome, and phenotype, and the role of the 4DN in cancer cells is not known. We have therefore assembled a team with diverse expertise spanning cell biology, computer science, and mathematics. The objective of this project is to determine the effect of perturbation of colorectal cancer cells on the 4DN of cancer. This will be accomplished by completing two specific aims in which we will perturb two colorectal cancer cell lines: 1) Quantify the impact of WNT signaling perturbation on the 4DN of human cancer. We will use RNA interference to attenuate TCF7L2, a key transcription factor in WNT signaling. The global impact of suppressed WNT signaling on the cancer 4DN will be assessed through genome-wide assays of architecture (Hi-C), transcription (RNA-seq), proteomics and super-resolution imaging, sampled through multiple time points. 2) Quantify the impact of aneuploidy perturbation on the 4DN of human cancer. We will alter aneuploidy with complementary approaches of silencing and artificially induced trisomy of Chromosome 13.